His Methodology is complete and utter crap. Company growth should be included, but also look at the long term strategy, customer satisfaction, employee satisfaction or employee turnover rate.
I think Brian Driscoll of Hostess needs to be on that list as well as Leo Apotheker, not because he was a CEO this year, but because he failed so bad at it last year.
Nobody gives a rat's rear about the long term. Get the most money you can as fast as possible, pass it out in fat raises and bonuses to the upper tiers, then watch them all flee for greener pastures.
No point in measuring for anyone's satisfaction or the company's long-term prosperity. That kind of stuff went out with the 20th Century. Not enough holdouts to be worth considering.
I was very specifically told that that was my problem and that I should be more like "Joe" who got stuff "done" faster because he didn't think things through or put in exception-handling logic. That it was, in fact "better" to reboot the server daily than to "waste time" making an app that cleaned up after itself. They were very big on "Done" in that particular shop.
I'm guessing that particular company sold the code to other companies and didn't actually use it themselves? That's the only (immoral) way I can see justifying such an attitude.
Unfortunately, no. They are a bank. A rather large one.
A companies first and foremost responsibility is to it's customers, 2nd to it's employees and finally 3rd to it's shareholders.
This bears repeating. Bankruptcy solves this problem as much as anything can. Perhaps due to the systemic risk of so many institutions failing at once, there should have been some sort of action to provide liquidity or whatever. But that could have been done on the cheap while the company was going through bankruptcy. Instead, they turned a crisis into a milk run, transferring wealth from the public to the people who caused the crisis.
You forgot creditors. They, too, have a specific place in line when a company goes bankrupt and it's higher than most everyone else.
Legally, yes. In many other respects, as a matter of long-term profitability, no. Take care of your customers and employees, and they will take care of you. Abuse them, and they will find someone else.
Regrettably, however, both employees and customers will tolerate a considerable amount of abuse these days.
"just Git 'R Dun!" which I think means just get it done, is a good way to dig yourself a nice big hole, you have to get it done right, surprisingly this turns out to be the fastest solution in most cases anyway since you don't have deal with the edge cases that so called quick solution created.
Unfortunately, my reputation IS for doing things correctly and carefully. I was very specifically told that that was my problem and that I should be more like "Joe" who got stuff "done" faster because he didn't think things through or put in exception-handling logic. That it was, in fact "better" to reboot the server daily than to "waste time" making an app that cleaned up after itself. They were very big on "Done" in that particular shop.
I used to always delete dead code immediately. I've backed off on that lately because occasionally it turns out that the cure was worse than the disease and it's good to have the previous-generation code available for immediate reference. Especially when I have the offending module in dead-tree form far from a differencing server. And I'm afraid that in any event, the version-compare tools available to me have a very definite latency - I can't just hover a mouse and have the changes immediately pop up in milliseconds. Instead I have to explicitly issue a set of secondary commands, wait for the differences to be generated and the UI components to initialize and scroll as needed. Not a LONG delay mind you, but long enough to discourage casual usage.
No, it is *VERY* concise. Specific words have an extremely specific meaning, and new concepts are defined in detail. That's why it reads the way it does. Writing it in plain english would take 5 or 6 times the amount of space to mean the same thing.
I've always maintained that the reason why Legal-sized paper is longer than "standard" (letter) paper is that the extra verbiage required extra space in order to fit on a single page. The legal advantage to putting stuff on a single page is that you can't surreptitiously swap out a page in a document if there's only one page to begin with.
OK, so it's one page, front-and-back, 4-point type, in gray ink on yellow paper, but still...
I tend to heavily bookmark those books and use the index, glossary and appendixes a lot, which is where e-readers don't have good solutions yet.
Actually, the Nook does those things pretty well, plus can do free text searches, which is not easily done when it means you have to skim every page in a dead-tree book.
What e-readers DON'T do well is handle documents with big diagrams or multiple detailed diagrams per page. Not enough screen space. Even if the resolution can handle it, a 7-inch version of some stuff I've seen can be murder on the eyes.
Also, I learn a lot via bibliomancy: picking up a technical book or magazine, letting it fall open to random topics, which I then may find worth reading. E-readers don't do that.
Full-sized bluetooth keyboards exist. I've seen both classic (rigid) and flexible models, no foldables though.
"Portable Stowaway". Couple of years old, folds up into a 4x5-inch pocket-sized package. A tad fragile, but otherwise a nice unit. I used to use it as the keyboard for my phone before I got a phone that didn't support Bluetooth keyboards.
The problem with numbers like that is that they imply a false precision. A 2x4, for example, isn't really all that close to being 2 inches by 4 inches, but because they are nice round numbers, people will cut them some slack. 70cm, on the other hand, implies precision to a small fraction of an inch - perhaps too small.
Despite ongoing efforts, I still have trouble relating degrees Celsius to "real" temperatures. For a long time, optimal temperature for me was 83 degrees F. 82 was "cool", 84 was "warm". 83 was perfection. I live in an area where 20 Celsius means that either I'm in an air-conditioned building or it's Winter, so that's fine, but nobody gives weather reports to the decimal place that I know of.
The metric system is more convenient for doing math in. The American/English system is more convenient for doing living in.
I can usually barely feel a 1-degree F difference in temperature. A 1-degree C difference, however, is almost twice that much, and that I can feel. So Fahrenheit is more practically precise, even if less aligned with physics.
Likewise, inches are slightly more convenient than centimeters, in large part because they are based off nominal body-part sizes and other common everyday objects instead of an abstruse mathematical derivative of the size of the Earth. More to the point, a lot of convenient distances are easier to quote. Like a 2-foot wide door instead of a "2/3 meter wide" door (approximately).
I never got properly corrupted by weights and fluid measures, so I am fine with either one there. A liter is close enough to a quart that it breaks down to pints and cups without too much inaccuracy for most of my needs. And weight is just something I read off a scale, so one number is as good as another for the most part. Especially when the number is based off another commonly-encountered thing (water).
Once I get into more esoteric realms, such as electromagnetism, there was never really much competition for units to begin with, so no problem.
The biggest reason that metric hasn't taken over in the USA is probably too many anti-flouridation folks who consider it as part of the Conspiracy to establish One World Order. But it must be popular with someone. Back in the 1980s (give or take), the State of Florida put up "88km/h" signs on the interstate below the 55mph speed limit signs. They were promptly stolen right and left.
A decimal system of weights and measures based on the meter and on the kilogram
kg are not a unit of weight. Newton is the unit of weight in SI.
You don't even use the correct SI units. Again Metric is not SI.
Darned if I'm going to let you out-pedant me. Just because a system of weights and measures is based on the meter and kilogram doesn't actually say the the kilogram itself is the measure of weight, just the basis for the measure of weight. So there!
Even less. Perhaps you have recieved a "proxy vote" letter, where they tell you what and who you are going to vote for during the upcoming meeting?
Yes. But that's only how "you" will vote if you don't explicitly vote your proxy. The proxy is simply a way of registering votes without having to attend the actual shareholder meeting. The only insidious thing about that is that if you don't choose, the choice will be made for you.
Most shareholder meetings have some sort of provision where shareholders can introduce more or less whatever issues they want to be voted on as long as they actually attend the meeting. Some companies even hold meetings in inconvenient places specifically to limit that.
But regardless, it's still one share, one vote when the votes get tallied. If you don't have the shares, your vote may be a spit in the swimming pool - a statement, perhaps, but not a decision.
I had two friends in the last 4 years fired the week before X-Mas. I mean, seriously, who the fuck comes up with that idea?
The Shareholders.
Another silly person who thinks that because corporations have elections, that makes them democracies.
It's not one person one vote, it's one share, one vote. Most corporations have the bulk of their shares concentrated in the hands of a few individuals and/or organizations. So of all the various parties interested in one way or another in a corporation: employees, customers, shareholders, vendors, the ones whose votes count are almost always a scant, narrow-minded minority.
In any event, when was the last time that you heard layoffs put up for a vote? The decision to Grinch the employees is made by someone who answers only in general terms, not on specific issues, and who answers to - as mentioned - a scant like-minded minority. None of whom are likely to have to cancel Christmas themselves. When members of that little clique fail, they get paid more than most people will get paid for a lifetime of success.
If Microsoft demands it, motherboard makers will fall in line in order to stay in business.
This probably also signals the beginning of the end of Microsoft.
Mighty empires always fall.
There's wholesale motherboards and retail motherboards. Wholesale motherboards are mostly destined for name-brand computers where MS-Windows will be pre-bundled.
However, when you buy retail, I'd venture that a lot of those motherboards have to be Linux-friendly, because Windows doesn't come "free" with them the way it does with mass-market computers and therefore I'd expect a much higher percentage of such motherboards to be destined for non-Windows machines, and since I have grave doubts about them becoming Apple machines, that leaves Linux as pretty much the largest market left.
In any event, so far Asus, Shuttle, MSI and BioStar have all worked fairly well for me. Occasionally an integrated peripheral will be problematic, but as far as it goes, I really wouldn't expect top-of-the-line integrated peripheral support from a retail mobo even on Windows. Especially considering what the Windows device driver development process has become.
Einstein was a mathematical prodigy. We preserved his brain for post-mortem study and it is unquestionably different from the average brain.
How much was the thinking shaping the brain and how much was the brain shaping the thinking isn't something we can determine, but his brain was definitely different. And, like Ramanujan, he had a reputation for never ceasing to think.
Which is fine, until a client sends you a document from MS Office and wants you to send back your changes with change tracking turned on, so that they can see what has changed in the document. If you only use it for internal documents, Google Docs can be fine. However, once you want to communicate with the outside world, you had better have MS office, or things will break down quite quickly.
When you want to communicate with an 800-lb. gorilla, you speak the language of the 800-lb. gorilla. If you want to do business with IBM, you use IBM's document formats. If you want to do business with Google, use Google formats. When IBM and Google want to do business with each other, they can either play document tag or agree on a common format.
The one question that nobody's asking here, is "Do they really mean explicitly MS Excel?" Or do they mean Excel as in "Kleenex"? If they're really just asking for spreadsheet proficiency and genericizing a brand name, we're getting all excited over the wrong things.
Reminds me of the time when I was younger that I applied for a senior engineer position. Granted, I wasn't really what you might call senior at the time, but the guy they ended up hiring was only senior in the sense that he could have joined the AARP.
Now, I have nothing against people who have been working for awhile, but this was in 1999, where his experience with internet technologies was probably about the same (or worse) than mine was, despite his 20 years or more in "technology". That was the funny thing about then and now. Today, you could have over a decade in Java programming and it is still in use. In the late nineties or 2000 or so, a recent college grad knew about as much about Java as someone who had been programming in other languages for a decade or more. You could argue that their experience was still useful, but I'd say it was a lot less useful, pound for pound, than senior level programming experience is today. That is, if you're using Java. If you stuck with coding in C/C++, your experience today is probably Godlike, assuming that your arteries haven't started to harden.
Not the best example. In 1999, Java was still mostly a C/C++ sibling language to the point that the SCJP questions tended to primarily focus on catching people who answered with C/C++ answers to Java-specific features. The main differences had to to with things like the standard class library and the interface contract form.
It really wasn't until after 2000 that Java began to develop its own personality, fleshing out the programming-by-interface, Inversion of Control, logging, Junit testing and the various enriched class packages from Sun, Apache, and others. Then again, some of these features also began to appear in C and C++, as well as in less java-like environments, so an experienced programmer these days needs a lot more breadth in the area of standard pre-constructed mechanisms than a 20th-Century programmer did, regardless of what language is being used.
Mo, of course not. But guess what? Insurance companies hire these people call - get this - Insurance Adjusters, who are - get this - "professionals" at evaluating claims.
But you know what? Most of them only look for reasons *NOT* to pay out on legitimate claims.
Well, either I've been very fortunate (more than once, alas), you're buying into the ambulance-chasing lawyer commercials, or you're getting insurance from a company whose "Lower Price Everyday[TM]" comes from making what you pay on difficult to collect. Because the adjusters I've dealt with have worked with me to get a decent job done even when some of the alternatives would have been cheaper. So they're not all worthless parasites, at least.
Slashdot Mirror
tl;dr short version: "A little knowledge is a dangerous thing."
People who think they know what the job entails start out saying "It's Easy! All You Have To Do Is..." and the whole thing swiftly descends into Hell.
His Methodology is complete and utter crap. Company growth should be included, but also look at the long term strategy, customer satisfaction, employee satisfaction or employee turnover rate.
I think Brian Driscoll of Hostess needs to be on that list as well as Leo Apotheker, not because he was a CEO this year, but because he failed so bad at it last year.
Nobody gives a rat's rear about the long term. Get the most money you can as fast as possible, pass it out in fat raises and bonuses to the upper tiers, then watch them all flee for greener pastures.
No point in measuring for anyone's satisfaction or the company's long-term prosperity. That kind of stuff went out with the 20th Century. Not enough holdouts to be worth considering.
Because some people have a sense of dignity and object to being treated like cattle.
Sale at Wal-Mart!
I was very specifically told that that was my problem and that I should be more like "Joe" who got stuff "done" faster because he didn't think things through or put in exception-handling logic. That it was, in fact "better" to reboot the server daily than to "waste time" making an app that cleaned up after itself. They were very big on "Done" in that particular shop.
I'm guessing that particular company sold the code to other companies and didn't actually use it themselves? That's the only (immoral) way I can see justifying such an attitude.
Unfortunately, no. They are a bank. A rather large one.
Give a man a fish and you feed him for a day.
Teach a man to fish and you destroy a whole ecosystem.
Dynamite!
Teach a man to fish etc
... when YOU own the only lake!
You bastard.
A companies first and foremost responsibility is to it's customers, 2nd to it's employees and finally 3rd to it's shareholders.
This bears repeating. Bankruptcy solves this problem as much as anything can. Perhaps due to the systemic risk of so many institutions failing at once, there should have been some sort of action to provide liquidity or whatever. But that could have been done on the cheap while the company was going through bankruptcy. Instead, they turned a crisis into a milk run, transferring wealth from the public to the people who caused the crisis.
You forgot creditors. They, too, have a specific place in line when a company goes bankrupt and it's higher than most everyone else.
Legally, yes. In many other respects, as a matter of long-term profitability, no. Take care of your customers and employees, and they will take care of you. Abuse them, and they will find someone else.
Regrettably, however, both employees and customers will tolerate a considerable amount of abuse these days.
"just Git 'R Dun!" which I think means just get it done, is a good way to dig yourself a nice big hole, you have to get it done right, surprisingly this turns out to be the fastest solution in most cases anyway since you don't have deal with the edge cases that so called quick solution created.
Unfortunately, my reputation IS for doing things correctly and carefully. I was very specifically told that that was my problem and that I should be more like "Joe" who got stuff "done" faster because he didn't think things through or put in exception-handling logic. That it was, in fact "better" to reboot the server daily than to "waste time" making an app that cleaned up after itself. They were very big on "Done" in that particular shop.
I used to always delete dead code immediately. I've backed off on that lately because occasionally it turns out that the cure was worse than the disease and it's good to have the previous-generation code available for immediate reference. Especially when I have the offending module in dead-tree form far from a differencing server. And I'm afraid that in any event, the version-compare tools available to me have a very definite latency - I can't just hover a mouse and have the changes immediately pop up in milliseconds. Instead I have to explicitly issue a set of secondary commands, wait for the differences to be generated and the UI components to initialize and scroll as needed. Not a LONG delay mind you, but long enough to discourage casual usage.
Yes, one can and should dig through old versions of the source. Your employer paid for it, you might as well use it. Like, um, duh!
My employer told me in no uncertain terms to "just Git 'R Dun!" He doesn't want me spending a lot of time looking at how things USED to work.
If a chunk of dead code is still present and commented out, I can scan it and go on. Very few of those fancy tools can give me feedback that fast.
thats what git is for. ;)
branches are cheap - use them
Merging branches, on the other hand, can be a royal pain.
No, it is *VERY* concise. Specific words have an extremely specific meaning, and new concepts are defined in detail. That's why it reads the way it does. Writing it in plain english would take 5 or 6 times the amount of space to mean the same thing.
I've always maintained that the reason why Legal-sized paper is longer than "standard" (letter) paper is that the extra verbiage required extra space in order to fit on a single page. The legal advantage to putting stuff on a single page is that you can't surreptitiously swap out a page in a document if there's only one page to begin with.
OK, so it's one page, front-and-back, 4-point type, in gray ink on yellow paper, but still...
I tend to heavily bookmark those books and use the index, glossary and appendixes a lot, which is where e-readers don't have good solutions yet.
Actually, the Nook does those things pretty well, plus can do free text searches, which is not easily done when it means you have to skim every page in a dead-tree book.
What e-readers DON'T do well is handle documents with big diagrams or multiple detailed diagrams per page. Not enough screen space. Even if the resolution can handle it, a 7-inch version of some stuff I've seen can be murder on the eyes.
Also, I learn a lot via bibliomancy: picking up a technical book or magazine, letting it fall open to random topics, which I then may find worth reading. E-readers don't do that.
Full-sized bluetooth keyboards exist. I've seen both classic (rigid) and flexible models, no foldables though.
"Portable Stowaway". Couple of years old, folds up into a 4x5-inch pocket-sized package. A tad fragile, but otherwise a nice unit. I used to use it as the keyboard for my phone before I got a phone that didn't support Bluetooth keyboards.
The problem with numbers like that is that they imply a false precision. A 2x4, for example, isn't really all that close to being 2 inches by 4 inches, but because they are nice round numbers, people will cut them some slack. 70cm, on the other hand, implies precision to a small fraction of an inch - perhaps too small.
Despite ongoing efforts, I still have trouble relating degrees Celsius to "real" temperatures. For a long time, optimal temperature for me was 83 degrees F. 82 was "cool", 84 was "warm". 83 was perfection. I live in an area where 20 Celsius means that either I'm in an air-conditioned building or it's Winter, so that's fine, but nobody gives weather reports to the decimal place that I know of.
The metric system is more convenient for doing math in. The American/English system is more convenient for doing living in.
I can usually barely feel a 1-degree F difference in temperature. A 1-degree C difference, however, is almost twice that much, and that I can feel. So Fahrenheit is more practically precise, even if less aligned with physics.
Likewise, inches are slightly more convenient than centimeters, in large part because they are based off nominal body-part sizes and other common everyday objects instead of an abstruse mathematical derivative of the size of the Earth. More to the point, a lot of convenient distances are easier to quote. Like a 2-foot wide door instead of a "2/3 meter wide" door (approximately).
I never got properly corrupted by weights and fluid measures, so I am fine with either one there. A liter is close enough to a quart that it breaks down to pints and cups without too much inaccuracy for most of my needs. And weight is just something I read off a scale, so one number is as good as another for the most part. Especially when the number is based off another commonly-encountered thing (water).
Once I get into more esoteric realms, such as electromagnetism, there was never really much competition for units to begin with, so no problem.
The biggest reason that metric hasn't taken over in the USA is probably too many anti-flouridation folks who consider it as part of the Conspiracy to establish One World Order. But it must be popular with someone. Back in the 1980s (give or take), the State of Florida put up "88km/h" signs on the interstate below the 55mph speed limit signs. They were promptly stolen right and left.
A decimal system of weights and measures based on the meter and on the kilogram
kg are not a unit of weight. Newton is the unit of weight in SI.
You don't even use the correct SI units. Again Metric is not SI.
Darned if I'm going to let you out-pedant me. Just because a system of weights and measures is based on the meter and kilogram doesn't actually say the the kilogram itself is the measure of weight, just the basis for the measure of weight. So there!
Even less. Perhaps you have recieved a "proxy vote" letter, where they tell you what and who you are going to vote for during the upcoming meeting?
Yes. But that's only how "you" will vote if you don't explicitly vote your proxy. The proxy is simply a way of registering votes without having to attend the actual shareholder meeting. The only insidious thing about that is that if you don't choose, the choice will be made for you.
Most shareholder meetings have some sort of provision where shareholders can introduce more or less whatever issues they want to be voted on as long as they actually attend the meeting. Some companies even hold meetings in inconvenient places specifically to limit that.
But regardless, it's still one share, one vote when the votes get tallied. If you don't have the shares, your vote may be a spit in the swimming pool - a statement, perhaps, but not a decision.
I had two friends in the last 4 years fired the week before X-Mas. I mean, seriously, who the fuck comes up with that idea?
The Shareholders.
Another silly person who thinks that because corporations have elections, that makes them democracies.
It's not one person one vote, it's one share, one vote. Most corporations have the bulk of their shares concentrated in the hands of a few individuals and/or organizations. So of all the various parties interested in one way or another in a corporation: employees, customers, shareholders, vendors, the ones whose votes count are almost always a scant, narrow-minded minority.
In any event, when was the last time that you heard layoffs put up for a vote? The decision to Grinch the employees is made by someone who answers only in general terms, not on specific issues, and who answers to - as mentioned - a scant like-minded minority. None of whom are likely to have to cancel Christmas themselves. When members of that little clique fail, they get paid more than most people will get paid for a lifetime of success.
If Microsoft demands it, motherboard makers will fall in line in order to stay in business.
This probably also signals the beginning of the end of Microsoft.
Mighty empires always fall.
There's wholesale motherboards and retail motherboards. Wholesale motherboards are mostly destined for name-brand computers where MS-Windows will be pre-bundled.
However, when you buy retail, I'd venture that a lot of those motherboards have to be Linux-friendly, because Windows doesn't come "free" with them the way it does with mass-market computers and therefore I'd expect a much higher percentage of such motherboards to be destined for non-Windows machines, and since I have grave doubts about them becoming Apple machines, that leaves Linux as pretty much the largest market left.
In any event, so far Asus, Shuttle, MSI and BioStar have all worked fairly well for me. Occasionally an integrated peripheral will be problematic, but as far as it goes, I really wouldn't expect top-of-the-line integrated peripheral support from a retail mobo even on Windows. Especially considering what the Windows device driver development process has become.
The man was a mathematical prodigy.
Einstein was a mathematical prodigy. We preserved his brain for post-mortem study and it is unquestionably different from the average brain.
How much was the thinking shaping the brain and how much was the brain shaping the thinking isn't something we can determine, but his brain was definitely different. And, like Ramanujan, he had a reputation for never ceasing to think.
Which is fine, until a client sends you a document from MS Office and wants you to send back your changes with change tracking turned on, so that they can see what has changed in the document. If you only use it for internal documents, Google Docs can be fine. However, once you want to communicate with the outside world, you had better have MS office, or things will break down quite quickly.
When you want to communicate with an 800-lb. gorilla, you speak the language of the 800-lb. gorilla. If you want to do business with IBM, you use IBM's document formats. If you want to do business with Google, use Google formats. When IBM and Google want to do business with each other, they can either play document tag or agree on a common format.
The one question that nobody's asking here, is "Do they really mean explicitly MS Excel?" Or do they mean Excel as in "Kleenex"? If they're really just asking for spreadsheet proficiency and genericizing a brand name, we're getting all excited over the wrong things.
Reminds me of the time when I was younger that I applied for a senior engineer position. Granted, I wasn't really what you might call senior at the time, but the guy they ended up hiring was only senior in the sense that he could have joined the AARP.
Now, I have nothing against people who have been working for awhile, but this was in 1999, where his experience with internet technologies was probably about the same (or worse) than mine was, despite his 20 years or more in "technology". That was the funny thing about then and now. Today, you could have over a decade in Java programming and it is still in use. In the late nineties or 2000 or so, a recent college grad knew about as much about Java as someone who had been programming in other languages for a decade or more. You could argue that their experience was still useful, but I'd say it was a lot less useful, pound for pound, than senior level programming experience is today. That is, if you're using Java. If you stuck with coding in C/C++, your experience today is probably Godlike, assuming that your arteries haven't started to harden.
Not the best example. In 1999, Java was still mostly a C/C++ sibling language to the point that the SCJP questions tended to primarily focus on catching people who answered with C/C++ answers to Java-specific features. The main differences had to to with things like the standard class library and the interface contract form.
It really wasn't until after 2000 that Java began to develop its own personality, fleshing out the programming-by-interface, Inversion of Control, logging, Junit testing and the various enriched class packages from Sun, Apache, and others. Then again, some of these features also began to appear in C and C++, as well as in less java-like environments, so an experienced programmer these days needs a lot more breadth in the area of standard pre-constructed mechanisms than a 20th-Century programmer did, regardless of what language is being used.
Ouch! Serves me right for not reading the original assertion more carefully.
Mo, of course not. But guess what? Insurance companies hire these people call - get this - Insurance Adjusters, who are - get this - "professionals" at evaluating claims.
But you know what? Most of them only look for reasons *NOT* to pay out on legitimate claims.
Well, either I've been very fortunate (more than once, alas), you're buying into the ambulance-chasing lawyer commercials, or you're getting insurance from a company whose "Lower Price Everyday[TM]" comes from making what you pay on difficult to collect. Because the adjusters I've dealt with have worked with me to get a decent job done even when some of the alternatives would have been cheaper. So they're not all worthless parasites, at least.